• State
  • Locked Locked
  • Replies 4 replies
  • Subscribers 30 subscribers
  • Views 1155 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

OOP PCB layout question: Shield of top side circuitry from sensing liquid and CIN1 stuck at maximum value

Trevor Lynch-Staunton
Prodigy 120 points
Other Parts Discussed in Thread: TIDA-00317, FDC1004

I have built a PCB similar to the design in your OOP paper TIDA-00317.  The Cin1 sensor size is 8" x 0.31", and the reference sensors are  0.3" x 0.31".  

The main difference in my design is that I placed all the circuitry including power supply, I2C transceiver, and the FDC1004 in the area between the sensing pads and the reference pads.  

1.) I used Shield 1 to isolate below the circuitry, is this the right choice?  Should I have used GND? I thought I might exceed the 400pF max on Shields then.  Maybe I shouldn't have put any copper plane beneath the ICs?  The area that I mention is the dark grey in the PCB 

Top Side

2.) I can measure a result from both reference sensors that seems reasonable, but when I measure from CIN1, I get a 7FFFFF reading, maximum reading.  I have cut the CIN1 trace and now get some value just below zero.  Any thoughts?  All CIN1, CIN2, and CIN3 are differential measurements against CIN4.

  • 0
    TI__Expert 7400 points

    Hi:

    Can you give me some data so I can see what the response looks like as you fill the container with liquid at 1 unit intervals? Are you getting maximum readings even when it's empty. If that's the case or if it reads max when you start filling it with liquid, then your total capacitance is larger than the input capacitance range of the device.

    There are a few ways that you can offset it since you cannot use the CAPDAC. One way is to put a fixed capacitor on CH4, but the issue now is that CH1, CH2, CH3 are affected by this, so you may find yourself being outside of the input cap range for CH2 or CH3.

    Also, you could have used GND but the issue would be a significant reduction in sensitivity compared to the SHLD. You do not need to make the SHIELD or GND that protects the circuitry the height of the container, you can make it a little higher than where the circuitry ends.

    Thanks

    -David Wang

    Capacitive Sensing Applications

  • 0 in reply to David Wang11
    Prodigy 120 points
    Thanks David for the prompt reply,
    Unfortunately, the reading is max when the container is empty. My dimensions are not much different than your TIDA-00317 design so I am not sure why that is the case. My flex circuit insulator material is only 2mil thick with 0.5oz copper on both sides. Perhaps that is much thinner than yours from the reference design? I cannot think of any other reasons.

    Even without the sensor placed on the container and just floating in space it reads max. Adding a cap to Cin4 could help.

    I will try to use an impedance analyzer to determine the overall capacitance of the sensing pad, but my impedance analyzer is 1Mhz to 1GHz. What should the reference for this measurement be?
  • 0 in reply to Trevor Lynch-Staunton
    TI__Expert 7400 points
    Since SHLD2 is out of phase of CIN1, with your flex circuit being thinner, it could have some larger parasitic cross coupling with the SHLD2 on the backside of SHLD2. Also, with the height of your sensors being more than twice the height of what I used, it could also be causing larger parasitic capacitance. Did you originally prototype these sensor dimensions with like copper tape prior to building the flex circuit?

    You can use an impedance analyzer, just be careful about the capacitance on the line since that will also affect what your "true" capacitance is. The excitation signal itself is 25kHz.
  • 0 in reply to David Wang11
    Prodigy 120 points
    Sorry this is the first prototype. I developed the ability to manufacture my own proto flex PCB, using a laser, etching, and some double sided Pyralux. Not the easiest, but worked very well.

    For a quick experiment, I will cut off my sensors and try some copper tape at various lengths tacked in place. I will let you know what I find out.